@inproceedings{tekiroglu-etal-2022-using,
title = "Using Pre-Trained Language Models for Producing Counter Narratives Against Hate Speech: a Comparative Study",
author = "Tekiro{\u{g}}lu, Serra Sinem and
Bonaldi, Helena and
Fanton, Margherita and
Guerini, Marco",
editor = "Muresan, Smaranda and
Nakov, Preslav and
Villavicencio, Aline",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2022",
month = may,
year = "2022",
address = "Dublin, Ireland",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-acl.245",
doi = "10.18653/v1/2022.findings-acl.245",
pages = "3099--3114",
abstract = "In this work, we present an extensive study on the use of pre-trained language models for the task of automatic Counter Narrative (CN) generation to fight online hate speech in English. We first present a comparative study to determine whether there is a particular Language Model (or class of LMs) and a particular decoding mechanism that are the most appropriate to generate CNs. Findings show that autoregressive models combined with stochastic decodings are the most promising. We then investigate how an LM performs in generating a CN with regard to an unseen target of hate. We find out that a key element for successful {`}out of target{'} experiments is not an overall similarity with the training data but the presence of a specific subset of training data, i. e. a target that shares some commonalities with the test target that can be defined a-priori. We finally introduce the idea of a pipeline based on the addition of an automatic post-editing step to refine generated CNs.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tekiroglu-etal-2022-using">
<titleInfo>
<title>Using Pre-Trained Language Models for Producing Counter Narratives Against Hate Speech: a Comparative Study</title>
</titleInfo>
<name type="personal">
<namePart type="given">Serra</namePart>
<namePart type="given">Sinem</namePart>
<namePart type="family">Tekiroğlu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Helena</namePart>
<namePart type="family">Bonaldi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Margherita</namePart>
<namePart type="family">Fanton</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marco</namePart>
<namePart type="family">Guerini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2022</title>
</titleInfo>
<name type="personal">
<namePart type="given">Smaranda</namePart>
<namePart type="family">Muresan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Preslav</namePart>
<namePart type="family">Nakov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aline</namePart>
<namePart type="family">Villavicencio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Dublin, Ireland</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>In this work, we present an extensive study on the use of pre-trained language models for the task of automatic Counter Narrative (CN) generation to fight online hate speech in English. We first present a comparative study to determine whether there is a particular Language Model (or class of LMs) and a particular decoding mechanism that are the most appropriate to generate CNs. Findings show that autoregressive models combined with stochastic decodings are the most promising. We then investigate how an LM performs in generating a CN with regard to an unseen target of hate. We find out that a key element for successful ‘out of target’ experiments is not an overall similarity with the training data but the presence of a specific subset of training data, i. e. a target that shares some commonalities with the test target that can be defined a-priori. We finally introduce the idea of a pipeline based on the addition of an automatic post-editing step to refine generated CNs.</abstract>
<identifier type="citekey">tekiroglu-etal-2022-using</identifier>
<identifier type="doi">10.18653/v1/2022.findings-acl.245</identifier>
<location>
<url>https://aclanthology.org/2022.findings-acl.245</url>
</location>
<part>
<date>2022-05</date>
<extent unit="page">
<start>3099</start>
<end>3114</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Using Pre-Trained Language Models for Producing Counter Narratives Against Hate Speech: a Comparative Study
%A Tekiroğlu, Serra Sinem
%A Bonaldi, Helena
%A Fanton, Margherita
%A Guerini, Marco
%Y Muresan, Smaranda
%Y Nakov, Preslav
%Y Villavicencio, Aline
%S Findings of the Association for Computational Linguistics: ACL 2022
%D 2022
%8 May
%I Association for Computational Linguistics
%C Dublin, Ireland
%F tekiroglu-etal-2022-using
%X In this work, we present an extensive study on the use of pre-trained language models for the task of automatic Counter Narrative (CN) generation to fight online hate speech in English. We first present a comparative study to determine whether there is a particular Language Model (or class of LMs) and a particular decoding mechanism that are the most appropriate to generate CNs. Findings show that autoregressive models combined with stochastic decodings are the most promising. We then investigate how an LM performs in generating a CN with regard to an unseen target of hate. We find out that a key element for successful ‘out of target’ experiments is not an overall similarity with the training data but the presence of a specific subset of training data, i. e. a target that shares some commonalities with the test target that can be defined a-priori. We finally introduce the idea of a pipeline based on the addition of an automatic post-editing step to refine generated CNs.
%R 10.18653/v1/2022.findings-acl.245
%U https://aclanthology.org/2022.findings-acl.245
%U https://doi.org/10.18653/v1/2022.findings-acl.245
%P 3099-3114
Markdown (Informal)
[Using Pre-Trained Language Models for Producing Counter Narratives Against Hate Speech: a Comparative Study](https://aclanthology.org/2022.findings-acl.245) (Tekiroğlu et al., Findings 2022)
ACL